VOL. 43.
No. 3.
MARCH,1921.
THEJOURNAE OF THE
American Chemical Society with which has been incorporated the
American Chemical Journal (Founded by Ira Remsen)
TWENTY-SEVENTH ANNUAL REPORT OF COMMITTEE ON ATOMIC WEIGHTS. DETERMINATIONS PUBLISHED DURING 1920. BY GREGORY PAULBAXTER. Received February 9, 1921.
In the table adopted by the International Committee on Atomic Weights in its report for 1920,' the only change from the table of 1919 is in the atomic weight of scandium, from 44.1 to 45.1. This change is based on Honigschmid's work noted below. Carbon.-Timmermans2 has determined the density of propane. Two methods of preparing the gas were employed: (1) the action of normal propyl iodide upon sodium dissolved in ammonia; (2) the action of budyronitrile on sodium. Both specimens were purified by fractional
distillation. The following weights of the normal liter were obtained. Method of preparation.
1 ............................ l.............. .............. 1 ............................ 1 ............................ Average.. ....................
Globe I. 564.88 cc.
455.71
2.01989 2.02181 2.02038 2.01878 2.02021
2.01896 2.01775 2.01908 2.02078 2.01913
THISJOURNAL, 42,1761 (1920). Tinunexmans, J . chim. phys., 18,133 (1920).
Globe 11. CC.
Globe 111. 351.91 cc.
2.01712 2.02087 2.02087 2.02214 2.02025
3s4
C;RGCOKT P.\UL
BAXTER.
Globe I. 564.88 cc.
Method of preparation.
Globe 11. 455.77 cc.
(2,01666)
(2.01645) 3.02129 2....................... 2.01901 Average . . . . . . . . . . . . . . . . . . . . . . 2.01836 2.02015 Average of all determinations. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Globe 111. 351.91 cc.
(2.01858) 2.01915 2.02320 2.02117 2.01986
Provisional calculation of the molecular weight of propane from the critical constants gave values for the atomic weight of carbon varying from 12.033 to 12.080 according to the method used. The author gives little weight to these results because of uncertainty as to the value of the critical constants. Fluorine.-Moles and Batuecas' have determined the density of methyl fluoride. Two methods of preparation were used: (1) The action of potassium methyl sulfate on potassium fluoride; ( 2 ) the action of methyl iodide on silver fluoride. The product of both methods was fractionally distilled. In the following table are given the weights of the normal liter determined a t pressures in the neighborhood of one atmosphere. Method,of preparatlon.
Globe I. 793.68 cc.
Globe I L . 619.35 cc.
Globe 111. 220.26 cc.
1.54416 1.54286
1.54371
1.54353 1.54368
1.54782 1.54782 1.54586 1.54425
....
I............................
....... ...................... ... I1 . . . . . . . . . . . . . . . . . . . . . . . . . . .
......... I1 . . . . . . . . . . . . . . . . . . . . . . I1 . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................
1.54350 1.54192 1.54222 1.54478
.................. Average . . . . . . . . . . . . . . . . . . . . . . Average of all determinations.. . . . Corrected for error in barometric sc
....... ....... ....... 1.54732
....... 1.54600 1.54061 1.54629 1.54241 1.54520
.......
....... ....... 1.54600 1.54336 1.54231
.......
1.54492 ..... 1.54489 . . . . . . . . . . . . . . . . . . 1.54542
Determinations made a t pressures not far from two-thirds atmospheric gave the following results. Globe I.
Globe 11.
a
1.53452 1.53850 1 ,53633 1.53896 1.53652 1.53952 1.53534 1,53413 1.53723 1.53&17 . . . . . . . . . . . 1.53599 1.53792 rminations . . . . . . . . . . ........... Corrected for error in barometric scale.. . . . . . . . . . . . . . . . . . . . . Probably 1.53683.
1
Moles and Batuecas, J . chinz. PJzys., 17,537 (1919).
Globe 111.
1.53174 1.53142 1.53093 1.53338 1.63683" 1.53286 1.53559 1,53611
3%
R E P O R T OP COMMITTEE ON ATOMIC \VEIGHTS.
At approximately one-third atmosphere the following figures were obt ained .
Average . . . . . . . . . . . . . . . . . . . . . . Average of all determinations Corrected for error in barome
Globe I.
Globe 11.
1.52445 1.52669 1.52819 1.52644
1.52445 1.52519 1,52900 1.52621
Globe 111.
.......
1.52480 1.52846 1.52663 1.52640 1.52692
..................... ...................... From the 3 values of the weight of the normal liter the value of 1 X a t one atmosphere is computed to be 1.0181. Using as the weight of the normal liter of oxygen 1.42905 g. and as the deviation from Avogadro's rule 1.00097, the molecular weight of methyl fluoride is found t o be 34.024, and the atomic weight of fluorine 19.001 (C = 12.000; H = 3 .OOi7). The weight of the normal liter of oxygen was determined with the s,ame apparatus as that used in the methyl fluoride work. Globe I.
Globe 11.
Globe 111.
1.42854 1.42925
1.42930 1.42935 1,42887 1.42917
1,42870 1,42843 1.42827 1.42847 1.42884 1.42933
.......
Average 1.42890 Average of all determinations.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . Corrected for error in barometric scale..
..
+
Using the latter value for oxygen the molecular weight of methyl fluoride is computed to be 34.018, and the atomic weight of fluorine 18.995. The authors prefer the average value 18.998. This agrees as well as could be expected with Smith and van Haagen's recent work. Aluminum.-Richards and Krepelkal synthesized aluminum bromide from the purest bromine and aluminum containing only 0.3y0of metallic impurities. After digestion in a current of nitrogen i t was fractionally distilled in nitrogen and in a vacuum, the first and last fraction being discarded. The purified material was finally distilled into small sealed glass bulbs. Analysis by comparison with silver and silver bromide gave the following results. Weights are corrected to vacuum. Br = 79.916. Wt. of AIBrs.
Wt. of AgBr.
3.23784
6.83996
5.03798 5.40576 3.41815 1.98012
1
Wt. of Ag.
Ratio AlBr: : 3Ag.
Preliminary analysis. ....... 0.47337a Final analyses. 6.11324 0.824110 ....... 6,55955 0.824105 ....... 4.14786 0.824076 ....... 2.40285 0.824071 Average 0.824090 Weighted Average 0.824098
AlBrs: 3AgBr. Richards and Krepelka, THISJOURNAL, 42, 2221 (1920).
At. wt. Al.
26.944 26.967 26.965 26.956 26.954 26.960 26.963
This result is 0.5% lower than the value adopted by the International Committee some time ago, 27.1. Silicon.-Baxter, Weatherill and Holmes' fractionally distilled silicon tetrachloride, prepared by the action of chlorine on silicon, in a sealed and exhausted all-glass apparatus. Weighed glass bulbs containing the purified material were broken under sodium hydroxide solution and the solution was then compared with silver in the usual way. Vacuum weights are given. C1 = 35.457. Fraction of SiClr.
Wt. of SiClr.
Wt. of Ag.
Ratio Sic16 : 4Ag.
12 3 9 6
10.43530 5.97853 8.79053 6.83524
26.49523 15.18304 22.32131 17.35617
0.393856 0.393764 0.393814 0.393828
At. wt. Si.
28.129 28.089 28.112 28.114 Average 28.11 1
Scandium.-Honigschmid2 converted scandium material, which had been purified by Meyer and Schweig3 (I) and Sterba-Bohm (11) into bromide by sublimation in a current of bromine from a mixture of oxide and carbon. The sublimate was collected in the weighing tube without exposure t o moisture and after being weighed was compared with silver. Weights are in vacuum. Br = 79.916. SAMPLE I. Wt. of SCBri.
Wt. of Ag.
2.79839 3.32001 3.34311 3.22367 3.11816 2.32284 2.13467 1.96690
3.17972 3.77235 3.79829 3.66250 3.51255 2.63915 2.42519 2.23460
At. wt. Sc. 45.079 45.084 45.108 45.114 45.120 45,103 45.122 45.120
Average 45.105
SAMPLE11. Wt. of ScBrr.
3.13887 2.92675 3.14889 2.79650 2.80219 2.62351 2.59879 2.60299 2.37863 1.41354
Wt. of Ag.
At. wt. Sc.
3.56645 3.32529 3.57784 3.17719 3,18371 2.98093 2.95270 2.95768 2.70257 1.60630
45 091 45 103 45.090 45.113 45.108 45.087 45.100 45.080 45.099 45.051 I
I
Average 45.093 Baxter, Weatherill and Holmes, THIS JOURNAL, 42, 1194 (1920). 2 Honigschmid, 2. Elektrochem., 25, 93 (1919). * Meyer and Schweig, Z.anorg. Chrn., 108,313 (1919). 1
387
REPORT OF COMMITTEE OK ATOMIC WEIGHTS.
The average value 45.10 is a unit higher than the former International value. Tin.-The complete papers by Baxter and Starkweather,' by Brauner and Krepelka2 and by Krepelka3 have been published. The data have already been included in the last 2 reports. Tellurium.-Bruylants and Michielsen4 purified by fractional distillation tellurium hydride which had been prepared by electrolysis of sulfuric acid with a tellurium cathode. The gas was weighed in closed tubes and after decomposition and combustion of the hydrogen the water was collected and weighed. W t , of TeH2.
Wt. of HzO.
Ratio Te : 0.
1.95150 0.79827 0.49887
0.27096 0.11078 0.06945
7.9833 7.9872 7.962
wt. Te. 127.73 127.79 (127.4)
At.
Samarium.-Owens, Balke and Kremers6 purified samarium material by fractional crystallization of magnesium samarium double nitrate. After 180 series of crystallizations had been made bismuth nitrate was added and 190 additional series carried out. Extreme fractions were frequently rejected. Bismuth was removed as sulfide and the samarium converted to chloride through the oxalate and oxide. Finally the carefully dehydrated chloride was compared with silver. Weights are corrected t o vacuum. C1 = 35.46. Fraction.
Wt. of SaCla.
Wt. of Ag.
Ratio SaCIr : 3Ag.
At. wt. Sa.
42 42 43 43
6.47502 4.03877 4.25295 4.34537 3.84046 3.32317 2.96369 3 34818 3.27999 3.00394 2.36481 3 68616 3.17435 2.81639 3.13555 2.91019 4.15775 3 32505 2.06632
8.15810 5.08083 5.35928 5.47663 4 84002 4.18820 3.73504 4.21936 4 13333 3.78769 2.97970 4.64577 4.00020 3.54932 3.95155 3.66744 5.23973 4.18952 2.62950
0.793692 (0.794903) 0.793567 0.793442 0.793480 0.793460 0.793480 0.793527 0.793546 0.793396 0.793640 0.793444 0.793548 0.793501 0.793500 0.793520 0.793504 0.793650 0.793430
150.49 (150.90) 150.45 150.41 150.42 150.42 150.42 150.44 150.44 150.40 150.47 150.41 150.44 150.43 150.43 150.44 150.43 150.47 150.41
44 45 32 33 34 35 3G 37 38 39 40 41 42 43 4-1
1
4 6
Average 0.793518 150.43 Baxter and Starkweather, THISJOL-RYAL, 42, 905 (1920) Braumr and Krepelka, z b z d , 42, 917 (1920) Kreprlka, zbzd., 42, 922 (1920) Bruylants and Mchielsen, Bull. acad. belg., Classe Sci., 1919,p. 119. Owens, Balke and Kremers, THISJOURNAL, 42, 515 (1920).
385
GREGORY PAUL BAXTER.
The average result is almost identical with the value recently found by James and Stewart by the same method. Thulium.-James and Stewart' converted carefully fractionated thulium material2 to chloride. The salt was carefully dehydrated, and after being weighed, was compared with silver in the standard fashion. Fraction A is the purest. Fractions B and C were known to contain neoytterbium. Vacuum weights are given. C1 = 33.457. Fraction.
Wt. of TmCla.
Wt. of Ag.
A A
2.17052 4,01446
2.54671 4.71091
At. wt. Tm.
Ratio TmClr : 3Ag.
0,85228 0.85216
169.46 169.42
Average . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
0.85222
169.44
B B
0.85279 0.85299
169.63 169.69
0.85289 0.85365
169.66 169.90
2.03868 1.53851
2.39060 1 ,80368
Average . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C 2.35242 2,75573
The lowest value is markedly higher than the one in current use, 168.5. Radioactive Lead.-Richards and Sameshima3 using the chloridesilver method, compared ordinary lead with lead from a deposit in a hot spring in Formosa containing radioactive elements. Wt. of PbCl?.
Ordinary lead. . . . 3.13929 3.09476
Wt. of Ag.
Ratio PbClz : 2Ag.
At. wt. Pb.
2.43553 2.40100
1.28895 1 28894
207.185 207.183
Average . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Japanese lead. . . . . 2.16756 1.14536 1.34496
1.68212 0.88881 1 ,04358
1 ,28859 I ,28864 1.28879
207.184 207.11 207.12 207.15
Average . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
207.13
Thorium Lead.-HOnigschmid* has analyzed lead chloride prepared from lead extracted from Norwegian thorit of the following composition: Th, 30.1%; U, 0.45%; Pb, 0.35%. Wt. of PbCIi.
Wt. of Ag.
Wt. of AgCl.
2.31995 2.32639 2.31995
1.79545
....... 2.39194 2.38516
....... .......
At. wt. Pb.
207.88 207.90 207.92 Average 207.90
Lead was extracted also from 3 specimens of thorianite, probably from Ceylon. James and Stewart, THIS JOURNAL, 42, 2022 (1920). Ibid., 33, 1332 (1911). Richards and Sameshima, ibid., 42, 928 (1920). 4 Honigschmid, 2. Elektrochem., 2 5 , 91 (1919). This work was noted in the 26th annual report, but without the detailed data.
389
REPORT OF COMMITTEE OK ATOMIC WEIGHTS. Tho>. 70.
Pb.
UOr.
70. 11.8 20.2 26.8
I.................... 68.9 . . . . . . . . . .i . . 6 2 . 7 111. . . . . . . . . . . . . . . . . 5 7 . 0
70. 2.3 3.1 3.5
Sample.
Wt. of PbCla.
Wt. of Ag.
Wt. of AgC1.
At. wt. Pb.
I
3.55514 4.85116 2.70841 3.55514 4.85116
2.75795 3.76332 .......
.......
....... .......
3.66408 5.00011
207.21 207.21 207.18 207.24 207.22
3.08009 4.18353 2.39883 3.08009 4.13353
2.39231 3.24896
.......
....... ....... .......
3.50716 3.17832 4.31608
4.75163
3.69124
....
Average
I1
.......
Average
I11
207.21 206.88 206.91 206.91 206.90 206.96 206.91 206.83
The values to be expected from the relative proportions of uranium and thorium are: (I) 207.26-.43; (11) 206.75-.98; (111) 206. 79-.99. Bismuth.-Honigschmidl has redetermined the atomic weight of bismuth by analysis of the chloride. Several specimens of carefully purified bismuth metal were converted to chloride in an atmosphere of dry chlorine, and the salt, after sublimation in a current of nitrogen in a weighed quartz tube, was transferred to a weighing bottle without exposure t o moisture. Analysis by comparison with silver and silver chloride followed. Since it was necessary t o dissolve the salt in 3 N nitric acid to avoid hydrolysis, comparative experiments with potassium chloride were carried out to show that this procedure was not attended with loss of chlorine or occlusion of silver nitrate. Weights are reduced tovacuum. C1 = 35.457. Wt. of KCl.
W t . of AgC1.
At. wt. K.
6.78151 7.92696 7.97062
39.096 39.097 39.097
Acid.. . . . . . . . . . . . . . . . . . 3.52723 Acid . . . . . . . . . . . . . . . . . . . 4.12307 Neutral.. . . . . . . . . . . . . . . 4.14578 Preliminary series. Wt. of BiCla.
Wt. of AgC1.
Ratio BiCls : 3AgCI.
At. wt. Bi.
6.61496 4.75395 7.48790 5,20711 5,15091
0.733401 0.733350 0.733367 0.733420 0.733435
209.000 208.978 208.985 209.008 209.014
Total 21.42597 29.21483 0.733395 Honigschmid, 2 . EZektrochenz., 26, 403 (1920).
-308.997
Sample.
1 1 1 1 1
4.85142 3.48631 5.49138 3,81900 3,77786
___
390
GREGORY PAUL BAXTER.
Final series. Sample. 1 1 1 1 1 1 2 2 2 2 5 5 4b 2 Total Sample. 1 1 1 1 1 1 1 1 3 2 2 2 2 5 5 4 4h 2 Total
Wt. of Ag.
Ratio BiCla : 3Ag.
At. wt. Bi.
3.29894 3.54331 4.74125 2.64020 4.49476 5.19912 4.99471 5.29284 4.62983 5.67213 5.69460 5.96312 6.24054 6.30709
3.38522 3.63594 4.85523 2.70934 4.61203 5.33506 5.12842 5,43129 4.75076 5 305.3 5.84367 6 ,11923 6,40379 G 47219
0.974513 0.974524 0.974517 0.974481 0.974573 0.974519 0.974498 0.974509 0.974545 0.974504 0,974490 0.974489 0.974507 0.974491
209.020 209.024 209.022 209.010 .209.040 209.022 209.015 209.019 209.031 209.017 209.013 209.012 209.019 209.013
68.71244
70.50970
Wt. of BiCla
Wt. of AgC1.
Ratio BiCla : 3AgC1.
4,49789 4.83067 6.4645.5 3.59956 6.12841 7.08896 5.94742 6.02960 6.57976 6.80969 7.21632 6.31252 7 73383 7.76479 8 ,13087 7 .(iOG32 8.50911 8.60003
0,733442 0.733503 0.733423 0.733479 0.733430 0.733411 0.733461 0,733463 0.733408 0.733471 0.733434 0.733436 0.733418 0.733388 0.733393 0.733419 0.733395 0.73338r)
209,020 Bi. 209.022 209.043 209.009 209,033 209.012 209.004 209.025 209.026 209.003 209.030 209 014 209.015 209.007 208.994 208.996 209.007 208.997 208.930
119.85070
0.733431
209.013
Wt. of BiCls
3.29894 3.54331 4,74125 2.64020 4.49476 5.19912 4.36220 4.42249 4,82566 4.99471 5.29281 4.62983 5.67213 5.69460 5.96312 5.57877 6.24054 6.30709 87.90156
0.974508
A t . wt.
The rounded average, 209.02, is 1.02 unit, or almost 0.5y0higher than the value in use a t the present time, 208.0. Wourtzell discusses the application of the method of limiting densities to the determination of the molecular weights of easily liquefiable gases, with special reference to sulfur dioxide. Cambridge 38, Mass.
Wourtzel, J . chim. phys., 18, 150 (1920).
